The purpose of this research is to test long-standing rival hypotheses that offer explanations for why there is such a high prevalence of mutations that cause disease phenotypes among the Ashkenazi Jews. The approach integrates a theoretical and empirical approach. First a computer model of Ashkenazi demographic history is constructed from historical records. Genetical evolution is simulated and probabilities of obtained observed genetic data are evaluated with a Monte Carlo procedure. A likelihood method will used to distinguish among competing hypotheses and parameterizations of models. The empirical approach consists of surveying a large number of microsatellite loci, as well as major histocompatibility complex sequence variation in several populations thought to represent the various stages of Ashkenazi colonization of Europe. These two genetic datasets will be tested for the possibility of a genetic bottleneck. If a bottleneck is found and other data are consistent with a demographic model of low long-term effective size, this suggests that selection can be inefficient at removing deleterious mutations from small populations. However, should there be no evidence for a bottleneck, this suggests that recent population's expansion can greatly increase the frequency of deleterious mutations.